International Journal on Artificial Intelligence Tools

We consider a variation of the problem of combining expert opinions for the situation in which there is no ground truth to use for training. Even though we do not have labeled data, the goal of this work is quite different from an unsupervised learning problem in which the goal is to cluster the data. Our work is motivated by the application of segmenting a lung nodule in a computed tomography (CT) scan of the human chest. The lack of a gold standard of truth is a critical problem in medical imaging. A variety of experts, both human and computer algorithms, are available that can mark which voxels are part of a nodule. The question is, how to combine these expert opinions to estimate the unknown ground truth. We present the Veritas algorithm that predicts the underlying label using the knowledge in the expert opinions even without the benefit of any labeled data for training. We evaluate Veritas using artificial data and real CT images to which synthetic nodules have been added, providing a known ground truth.

Maximum likelihood method for estimating parameters of Bayesian networks (BNs) is efficient and accurate for large samples. However, the method suffers from overfitting when the sample size is small. Bayesian methods, which are effective to avoid overfitting, present difficulties for determining optimal hyperparameters of prior distributions with good balance between theoretical and practical points of view when no prior knowledge is available.

As described in this paper, we propose an alternative estimation method of the parameters on BNs. The method uses a principle, rooted in thermodynamics, of minimizing free energy (MFE). We define internal energies, entropies, and temperature, which constitute free energies. Especially for temperature, we propose a "data temperature" assumption and some explicit models. This approach can treat the maximum likelihood principle and the maximum entropy principle in a unified manner of the MFE principle. For assessments of classification accuracy, our method shows higher accuracy than that obtained using the Bayesian method with normally recommended hyperparameters. Moreover, our method exhibits robustness for the choice of introduced hyperparameters.

Information extraction is of paramount importance in several real world applications in the areas of business, competitive and military intelligence because it enables to acquire information contained in unstructured documents and store them in structured forms. Unstructured documents have different internal encodings, one of the most diffused encoding is the visualization-oriented Adobe portable document format (PDF). Although several sophisticated and indeed complex approaches were proposed, they are still limited in many aspects. In particular, existing information extraction systems cannot be applied to PDF documents because of their completely unstructured nature that pose many issues in defining IE approaches. In this paper the novel ontology-based system named XONTO, that allows the semantic extraction of information from PDF documents, is presented. The XONTO system is founded on the idea of self-describing ontologies in which objects and classes can be equipped by a set of rules named descriptors. These rules represent patterns that allow to automatically recognize and extract ontology objects contained in PDF documents also when information is arranged in tabular form. This way a self-describing ontology expresses the semantic of the information to extract and the rules that, in turn, populate itself. In the paper XONTO system behaviors and structure are sketched by means of a running example.

A cognitive map model provides a graphical representation of an influence network between concepts. One drawback of this model is that large cognitive maps are difficult to exploit and understand. This paper introduces an ontological cognitive map model that enables the designer to organize concepts in an ontology. On the one hand, this model provides an ontological influence mechanism that shows the influence from any concept of the ontology to any other according to the map. The map is then easier to exploit. On the other hand, the ontology is used for providing a synthetical view of a map. The map is then easier to understand.

e-Tourism is a tourist recommendation and planning application to assist users on the organization of a leisure and tourist agenda. First, a recommender system offers the user a list of the city places that are likely of interest to the user. This list takes into account the user demographic classification, the user likes in former trips and the preferences for the current visit. Second, a planning module schedules the list of recommended places according to their temporal characteristics as well as the user restrictions; that is the planning system determines how and when to realize the recommended activities. Having the list of recommended activities organized as an agenda (i.e. an executable plan), is a relevant characteristic that most recommender systems lack.

Latent class analysis is a popular statistical learning approach. A major challenge for learning generalized latent class is the complexity in searching the huge space of models and parameters. The computational cost is higher when the model topology is more flexible. In this paper, we propose the notion of dominance which can lead to strong pruning of the search space and significant reduction of learning complexity, and apply this notion to the Generalized Latent Class (GLC) models, a class of Bayesian networks for clustering categorical data.

GLC models can address the local dependence problem in latent class analysis by assuming a very general graph structure. However, The flexible topology of GLC leads to large increase of the learning complexity. We first propose the concept of dominance and related theoretical results which is general for all Bayesian networks. Based on dominance, we propose an efficient learning algorithm for GLC. A core technique to prune dominated models is regularization, which can eliminate dominated models, leading to significant pruning of the search space. Significant improvements on the model.

Tree encodings of programs are well known for their representative power and are used very often in Genetic Programming. In this paper we experiment with a new data structure, named straight line program (slp), to represent computer programs. The main features of this structure are described, new recombination operators for GP related to slp's are introduced and a study of the Vapnik-Chervonenkis dimension of families of slp's is done. Experiments have been performed on symbolic regression problems. Results are encouraging and suggest that the GP approach based on slp's consistently outperforms conventional GP based on tree structured representations.

Given a Boolean formula in conjunctive normal form (CNF), the Exact Satisfiability problem (XSAT), a variant of the Satisfiability problem (SAT), consists in finding an assignment to the variables such that each clause contains exactly one satisfied literal. Best algorithms to solve this problem run in ( for X3SAT). Another possibility is to transform each clause in a set of equivalent clauses for the Satisfiability problem and to use modern and powerful solvers (zChaff, Berkmin, MiniSat, RSat etc.) to find such truth assignment. In this paper we introduce three new encodings from XSAT instances to SAT instances that lead to a lot of structural information (equivalency gates and and gates) which is naturally hidden in the pairwise transformation. Some solvers (lsat,march_dl,eqsatz) can take into account this kinds of structural information to make simplifications as pretreatment and speed-up the resolution. Then we show the interest of dealing with the XSAT formalism by introducing an encoding of binary CSP and graph coloring problem into XSAT instances. Preliminary results on real-world binary CSP and graph coloring problem show the importance of exhibiting equivalencies for the XSAT problem.